Clinical and Histopathologic Features and Immunologic Variables in Patients with Severe Chilblains
AbstractIntroduction Abbreviations used in this article: ACR, American College of Rheumatology, ANA, antinuclear antibodies, chilblain LE, chilblain lupus, CTD, connective tissue disease, LE, lupus erythematosus, SLE, systemic lupus erythematosus. Chilblains are cutaneous inflammatory lesions commonly occurring during cold and humid periods, which typically consist of erythematous papules involving fingers, toes, nose, or ears. While chilblains, also called perniosis, usually resolve within a few days, they may display unusual features, such as a prolonged course or the development of blisters or ulcers (5,7). Chilblain lesions following a prolonged course call for the differential diagnosis between idiopathic chilblains and chilblains associated with connective tissue diseases (CTDs), mostly lupus erythematosus (LE), and may herald LE in some cases (3,4,9,12,14,16). This latter entity was described initially in 1888 by Hutchinson (9) as chilblain lesions occurring in patients affected with cutaneous or systemic LE, consequently denominated “chilblain lupus” (chilblain LE). Although this rare form of LE has been recognized for many years, few studies have investigated its clinicopathologic and immunologic features, and both the retrospective modalities of previous reports and the limited sets of patients analyzed did not validate criteria allowing the distinction between idiopathic chilblains and chilblain LE (3,4,12,16). Furthermore, the prevalence of the association with CTDs has not been investigated in patients affected with severe chilblains. In the work presented herein, we prospectively studied 33 patients affected with severe chilblains—that is, chilblain lesions following a prolonged course. We focused on the prevalence of the association with CTDs, and we attempted to identify useful criteria for the differential diagnosis of chilblains associated with CTDs and idiopathic chilblains, and to delineate the essential features of chilblain LE. Patients and Methods Patient population Between 1991 and 1999, 33 patients (28 female and 5 male patients) were consecutively enrolled in a monocentric study conducted in the Dermatology Department of the Hôpital Saint-Louis, Paris, France. Patients were referred by dermatologists for chilblains following an unusually prolonged course, defined as lesions persisting for more than 1 month, and were included after giving their informed consent. Chilblains were defined as coldinduced inflammatory papular lesions showing an acral topography. The diagnosis of chilblains was established by the referring physician and confirmed by 2 investigators. Study design All patients had a complete physical examination by a dermatologist at initial evaluation. The following clinical parameters were assessed: topography of skin lesions, duration of disease, associated functional signs, existence of acrosyndrome (such as Raynaud phenomenon or acrocyanosis), and associated signs of CTDs. Thus, in addition to assessing patients for clinical American College of Rheumatology (ACR) revised criteria for systemic lupus erythematosus (SLE) (8,17), we systematically looked for the presence of sicca syndrome, purpura, and livedo reticularis or racemosa. A complete cardiovascular clinical examination was performed in all cases, allowing exclusion of patients with atheromatous disease. Laboratory tests included a complete blood count; serum creatinine level; serum protein electrophoresis; antinuclear antibodies (ANA) by immunofluorescence with Hep-2 substrate; anti-double-stranded DNA, anticardiolipin, and antiphospholipid autoantibodies by enzyme-linked immunosorbent assay (ELISA); anti-Sm, -RNP, -Ro/SS-A, and -La/SS-B autoantibodies by immunodiffusion; cryoglobulinemia; total serum hemolytic complement (CH50) by standard hemolytic assay; serum complement components C3 and C4 by radial immunodiffusion; rheumatoid factor by nephelometry; and 24-hour urine protein excretion. Skin punch biopsies (4 mm) of chilblain lesions were performed in 23/33 patients enrolled in the study. Histopathologic analysis was carried out with light microscopy on paraffinembedded sections stained with hematoxylin-eosin-saffron. A frozen sample was used for immunofluorescence studies to search for evidence of lupus band and/or intrakeratinocytary staining, as previously described (6). The presence of histopathologic changes suggestive of chilblain LE (3,12,16) was systematically investigated in an independent manner by 2 pathologists blinded with respect to clinical and laboratory variables. Thus, in addition to the presence of dermal edema and the depth of the dermal lymphocytic infiltrate and its perisudoral pattern, the following features were assessed: epidermal atrophy, basal linearization, basal thickening, keratinocyte vacuolization, and keratinocyte necrosis. Three cohorts of patients were defined according to clinical and laboratory findings at initial evaluation. Patients who did not exhibit any of the ACR revised criteria for SLE or any clinical sign or laboratory abnormality of CTDs, such as livedo reticularis or racemosa, purpura, sicca syndrome, or positive detection of cryoglobulinemia or rheumatoid factor, were classified in a first group called idiopathic chilblains. Consistent with previous studies (1), some patients in this group were affected with Raynaud phenomenon. Patients showing 1–3 ACR revised criteria for SLE and/or any clinical or laboratory manifestations of CTDs were classified in a second group denominated atypical chilblains. Finally, patients who fulfilled at least 4 ACR revised criteria for SLE or who had histologically ascertained cutaneous discoid lupus lesions were included in a third group called chilblain LE. A long-term clinical follow-up was performed. Patients had regular clinical evaluations by the same investigator every 6 months and in the case of new clinical events. Laboratory investigations were repeated only in the case of clinical changes suggesting evolution toward CTDs. Statistical analysis Comparisons between groups were made with the Wilcoxon test for continuous variables and the Fisher exact test for categorical variables. Two-sided tests were computed and p values of 0.05 or less were considered statistically significant. The SAS software program (SAS institute, Cary, NC) was used. Results Demographic characteristics of patients and clinical features of chilblain lesions The median age was 30.0 years at the time of initial evaluation (range, 13–53 yr). Fourteen patients were Caucasian, 4 originated from Africa, 12 were North African, and 3 were Afro-Carribean (Table 1,Table 2). In 2 patients, SLE was diagnosed before the onset of the chilblains. Patient 28 developed cutaneous and articular lupus 19 years before the onset of chilblain lesions, while patient 29 had severe thrombocytopenia with hemorrhage 22 years before chilblains (see Table 2).TABLE 1: Demographic, clinical, and laboratory features of 11 patients with idiopathic chilblainsTABLE 2: Demographic, clinical, and laboratory features of patients with atypical chilblains and chilblain LEThe median duration between the onset of initial chilblain lesions and entry in the study was 43.7 months (range, 1–240 mo). Functional signs related to chilblains lesions, in most cases consisting of pruritus or burning or itching sensations, were observed consistently in all groups. Fingers were the predominant localization in all groups (Figure 1), being involved in 85% of cases, while lesions involving toes (42%), auricular (9%) (Figure 2), and/or nasal areas (6%) were also observed (see Tables 1 and 2).Fig. 1: Ulcerated idiopathic chilblains on the dorsum of a finger.Fig. 2: Violaceous papulonodular lesions involving the auricular area in a patient with idiopathic chilblains (same patient as in Figure 1).According to our classification, 11 patients had idiopathic chilblains, 10 had atypical chilblains, and 12 patients had chilblain LE. Patients without associated signs of CTDs: Idiopathic chilblains Persistence of chilblains beyond cold seasons, that is, during the end of spring and summer, was not observed in any patient in this group (see Table 1). An acrosyndrome was noted in 6/11 patients of this group, consisting of Raynaud phenomenon (1 case) and acrocyanosis (5 cases). Serum levels of gammaglobulins were normal in all patients in this group. Cutaneous immunofluorescence was performed in 9/11 patients and showed negative results in 6 cases, while a slight, nonspecific C3 immunoreactivity of the dermal vessels was observed in 3 cases (see Table 1). Histologic features found most often were a deep dermal, perisudoral lymphocytic infiltrate (100%), dermal edema (6/8, 75%), keratinocyte necrosis (5/8, 62.5%), and keratinocyte vacuolization (4/8, 50%) (Table 3, Figure 3).Fig. 3: Hematoxylin-eosin stain analysis of a skin biopsy taken from an idiopathic chilblain lesion. A. Alterations consist of a slight epidermal hyperplasia with parakeratosis, a moderate papillar edema, with a lymphocytic pericapillary infiltrate (hematoxylin-eosin stain × 100). B. High magnification emphasizes the dense perisudoral lymphocytic infiltrate (hematoxylin-eosin stain × 200).TABLE 3: Histologic analysis of skin biopsies in 23 patients with severe chilblainsPatients with associated signs of CTDs: Atypical chilblains Six of the 10 patients included in the atypical chilblains group experienced at least 1 clinical ACR revised criterion for SLE at the time of chilblain lesions (see Table 2). Photosensitivity was present in 3 patients, while polyarthralgia or polyarthritis was noted in 4 cases. In 4/10 cases, lesions persisted beyond cold seasons. An acrosyndrome was noted in 5 patients of this group, consisting of Raynaud phenomenon in 3 of them. Erythromelalgia was observed in 1 patient (see Table 2). All these patients except 1 had positive detection of ANA in their serum, and the serum level of anti-DNA autoantibodies was above the normal upper limit in 4 of them. One additional patient (Patient 12) exhibited livedo reticularis associated with the presence of anti-DNA and anticardiolipin antibodies (see Table 2). A diagnosis of primary Sjögren syndrome was established in Patient 13 of this group (see Table 2). In this female patient, the history of an infant born 3 years before the study with neonatal LE was retrospectively assessed. Cutaneous immunofluorescence was performed in 6 patients in this group. It revealed a positive lupus band in 2 patients who showed concomitant presence of ANA, while results were negative in 4 patients (see Table 2). A deep dermal lymphocytic infiltrate (6/8, 75%), dermal edema (4/8, 50%), and the presence of keratinocyte vacuolization (2/8, 25%) were predominant histopathologic features (see Table 3). Patients with ascertained diagnosis of LE: Chilblain LE All patients belonging to the chilblain LE group experienced at least 1 clinical ACR revised criterion for SLE at the time of chilblain lesions. Lupus discoid lesions were observed in 7/12 patients, occurring before the onset of chilblains in 4 cases, with a mean delay of 7.5 years (range, 2–19 yr), while discoid lupus lesions occurred concomitantly with chilblains in 3 cases. Eight patients presented with SLE. Poly-arthralgia or polyarthritis was noted in 7 cases, while 2 patients had evidence of neuropsychiatric lupus at the time of study: 1 of them had seizures (Patient 32), and the other had evidence of psychosis (Patient 31). This latter patient presented with livedo reticularis and showed a lupus anticoagulant and a false-positive VDRL test. Lupus nephritis (class II of the World Health Organization classification) was diagnosed by renal histology in 1 patient (Patient 30). The association with sicca syndrome was observed in 1 patient. Chilblain lesions persisted beyond cold seasons in 5 cases. Raynaud phenomenon was present in 4 patients, while erythromelalgia was observed in 1 additional case. Hematologic abnormalities, that is, anemia, leukocytopenia, or lymphocytopenia, were observed in 8 patients, 1 of whom had evidence of autoimmune hemolytic anemia with a positive Coombs test (Patient 27). Hypergammaglobulinemia, defined as a serum level of gammaglobulins exceeding 18 g/L, was noted in 7/12 patients. ANA were detected in the serum of 8/12 patients in this group, with concomitant detection of anti-DNA autoantibodies in 4 cases. Hypocomplementemia was present in 3/12 patients (see Table 2). Cutaneous immunofluorescence was performed in 7 patients in this group and revealed evidence of a lupus band in 3 of them. These 3 patients had positive detection of ANA in the serum at the time of the study, and 1 of them also exhibited intracytoplasmic keratinocyte staining, which was associated with the presence of anti-SS-A and anti-SS-B autoantibodies (see Table 2). Dermal edema (6/7, 86%), deep dermal lymphocytic infiltrate (4/7, 57%), and the presence of keratinocyte vacuolization (3/7, 43%) were the 3 dominant histopathologic features. No basal linearization feature could be observed in any biopsy (Figure 4, see Table 3).Fig. 4: Histopathologic section of a chilblain lupus lesion. A. Epidermal hyperplasia with parakeratosis, papillar edema, and lymphocytic infiltrate surrounding dilated capillaries. Note the lack of epidermal atrophy and lack of significant alteration of the dermal-epidermal junction (hematoxylin-eosin stain × 100). B. Minimal lymphocytic infiltrate involving the perisudoral areas (hematoxylin-eosin stain × 200).Comparative analysis of clinical, laboratory, and histologic features in the 3 groups The 3 groups did not differ significantly according to age or ethnic group. The proportion of female patients was significantly higher in the chilblain LE group compared with the idiopathic chilblain group (100% versus 63%, p = 0.037). None of the clinical variables assessed in the present study—that is, functional symptoms; the presence of ulcerations, hyperkeratosis, or atrophy; and the anatomic localization of chilblain lesions—was found to differ significantly between patients with idiopathic chilblains and those with chilblain LE. Persistence of chilblains beyond cold seasons was noted in half the patients belonging to the atypical chilblains and chilblain LE groups, but was never observed in patients in the idiopathic chilblains group (p = 0.037). The prevalence of hypergammaglobulinemia was higher in patients in the chilblain LE group than in those in the idiopathic chilblains group (7/12 versus 0/11; p = 0.005). Histologic features also were analyzed comparatively. The prevalence of a perisudoral cellular infiltrate was higher in patients with idiopathic chilblains than in those with chilblain LE (8/8 versus 2/7, p = 0.007) (see Figures 3 and 4). The prevalence of other histologic features did not differ significantly among the 3 groups. Follow-up The median follow-up of patients was 38 months (range, 2–100 mo). No manifestation of LE or of any other CTDs occurred during follow-up of patients in the idiopathic chilblains group (median follow-up, 32 mo; range, 5–100 mo). The median follow-up of patients from the atypical chilblains group was 31 months (range, 2–63 mo). Healing of chilblain lesions occurred within the first 2 months of antimalarial treatment in 4 of 6 patients, while spontaneous remission of lesions (2 patients) and evolution to SLE (2 patients, Patients 15 and 18) were observed in the remaining patients who did not receive specific therapy. Patient 18 developed discoid cutaneous lesions 2 months after the initial evaluation. Sixteen months later, this patient exhibited full-blown SLE characterized by polyarthritis, hypocomplementemia, and positive detection of ANA and anti-Sm autoantibodies. Further course was complicated with central nervous system involvement leading to hemiparesis. The second case of SLE occurred in Patient 15, 36 months after the initial evaluation and 48 months after the onset of chilblains. This patient exhibited polyarthralgia, polyadenopathy, asthenia, hypocomplementemia, and positive detection of ANA, anti-DNA, and anti-SS-A autoantibodies. The median follow-up of patients in the chilblain LE group was 42.5 months (range, 17–100 mo). One patient developed discoid lupus lesions under treatment with antimalarials and low systemic steroid therapy, 27 months after entering the study. Essential characteristics of chilblain lesions associated with LE The retrospective analysis of the characteristics of the 14 patients with an ascertained diagnosis of LE (12 patients who were diagnosed with LE at entry and 2 patients who were diagnosed during follow-up) allows a better definition of essential characteristics of chilblain LE (Table 4). The onset of chilblains during the third decade and the predominance of female patients (13 female/1 male) and of patients originating from African countries (7/14) were remarkable demographic characteristics. Chilblain LE lesions presented as papular purpuric or violaceous, painful, or pruriginous lesions involving fingers in all patients (see Figure 4), and involving toes in one-third of cases. The nose, knees, and elbows were chilblain lesions during spring or were present in half of patients. LE lesions were observed in of our patients. the patients had and Raynaud phenomenon was assessed in of them. manifestations of SLE included hemolytic anemia (1 neuropsychiatric and renal involvement (1 case) and were detected in most patients during the initial evaluation. Hematologic consisting of and/or were noted in patients with chilblain LE, while ANA anti-DNA anticardiolipin autoantibodies anti-SS-A and/or anti-SS-B antibodies and hypergammaglobulinemia were laboratory findings (see Table 4). A lupus band was detected by immunofluorescence in 4: Essential characteristics of patients with chilblain Hutchinson (9) first described chilblain LE as papular lesions involving the associated with LE, retrospective studies have been with this form of LE, and that chilblain lesions may in association with both systemic and cutaneous of LE the prevalence of CTDs in patients affected with severe chilblains, that is, chilblain lesions following an unusually prolonged course, has not been The results of the present study the prevalence of clinical and/or laboratory features a diagnosis of in these patients, of our patients exhibited clinical of and/or laboratory evidence of patients in the atypical chilblains group showed manifestations of lupus, and 1 showed an association with Sjögren of chilblains with CTDs other than LE, such as and purpura, have been These the that a search for CTDs in patients with severe chilblains. In the follow-up study herein, we also the for development of in patients with severe idiopathic chilblains, and the long-term course in patients affected with chilblains who showed clinical and/or laboratory signs of immunologic 2 patients initially included in the atypical chilblains group developed full-blown SLE during follow-up treatment with antimalarial while patient included in the idiopathic chilblains group developed LE during the study. These the for a long-term follow-up of patients affected with severe chilblains, mostly those showing laboratory evidence of immunologic patients may clinical manifestations of LE after the onset of chilblain lesions. study focused on chilblain lesions in patients affected with CTDs, mostly LE, display specific clinical and/or histopathologic features that may them from idiopathic chilblains. In our study, clinical feature of a chilblain was found to be The only significant criteria were the of lesions during seasons and the of the patient. chilblain LE lesions were mostly in female patients showing a course. to histopathologic a retrospective study the that chilblain LE lesions histopathologic changes as those observed in discoid LE the established a of and criteria for the diagnosis of chilblain LE, with for histopathologic criteria the limited of patients included in that study and its retrospective any of these In to the latter study, we did not evidence of discoid lupus histopathologic features, such as epidermal atrophy or basal vacuolization, in chilblain LE lesions of our patients. This be by the that in retrospective biopsies could have been taken from cutaneous discoid lupus lesions involving the the results of our study are in with those by (1), who showed evidence of a higher prevalence of a deep perisudoral infiltrate in idiopathic chilblains than in chilblain LE lesions. their and from the present study that histopathologic feature of chilblain LE lesions, and consequently that skin biopsy usually not to chilblain LE lesions from idiopathic chilblains. these findings not biopsies of chilblain lesions for the differential diagnosis between the idiopathic form and chilblain LE. histopathologic examination of specific cutaneous lupus lesions associated with chilblains for the diagnosis of associated lesions of discoid LE. The detection of a lupus band in lesions of chilblain LE that cutaneous immunofluorescence may be useful to chilblain LE from idiopathic chilblains. these immunofluorescence findings were with the positive detection of ANA in the serum, immunofluorescence study not in cases with positive detection of ANA in the Finally, we attempted to the essential features of chilblain LE. Thus, we compared the clinical and laboratory of the 14 patients with an established diagnosis of chilblain LE in the present study with those of patients in 4 studies (see Table 4). Chilblains may be observed in both cutaneous LE and SLE. In of our results and those of previous articular involvement predominant among manifestations of chilblain SLE other have been in patients with chilblain LE, such as renal involvement and psychosis study additional in this showing evidence of severe such as seizures and lupus in chilblain SLE. It the prevalence of neuropsychiatric in patients with chilblain LE from that observed in the population of patients with SLE, which has been to from to Laboratory findings for our patients with chilblain LE the prevalence of and/or lymphocytopenia, and of remarkable findings in our study were the presence of cryoglobulinemia in 2 patients and of anti-SS-A autoantibodies in 3 patients. This low prevalence of anti-SS-A autoantibodies in with the results of 2 previous studies but in with those of who the positive detection of these autoantibodies in the serum of all studied patients. Thus, that specific laboratory with LE, in to other of LE, such as LE, which associated with the presence of anti-SS-A autoantibodies be that chilblain lesions may herald LE. A complete physical examination and laboratory investigations the search for laboratory signs of SLE are at evaluation of patients affected with chilblains following a course. long-term follow-up of patients with idiopathic chilblains following a course while for those showing any clinical and/or laboratory suggestive of an Further studies are to delineate the for development of LE and the modalities of follow-up in patients with chilblains. We investigated 33 patients affected with chilblain lesions following a persisting course of more than 1 We focused on the of an connective tissue disease, mostly lupus erythematosus (LE), and we analyzed features of idiopathic chilblains compared with those of chilblain lesions associated with connective tissue disease. We also carried out a follow-up of patients. patients included in the study were of any clinical and/or laboratory abnormality suggestive of connective tissue disease, while 22 of 33 patients showed 1 or for connective tissue disease, and among them 8 had a diagnosis of systemic lupus erythematosus (SLE) established at initial evaluation on the American College of Rheumatology revised The analysis of patients with idiopathic chilblains and patients with chilblains associated with LE showed that female and of lesions beyond cold seasons were significantly associated with chilblain LE. Histopathologic studies of lesions did not features of LE in any but revealed a higher of a deep perisudoral infiltrate in idiopathic chilblains. In patients showing signs of connective tissue disease, positive cutaneous immunofluorescence was with the presence of antinuclear patients had an ascertained diagnosis of SLE with severe manifestations during follow-up, treatment with in both cases. Chilblains following a course may connective tissue disease, and patients affected with chilblains associated with autoimmune may severe SLE. long-term follow-up of these patients We are to and for referring patients, and to for and of the
https://doi.org/10.1097/00005792-200105000-00004